Please use this identifier to cite or link to this item: https://hdl.handle.net/10356/150354
Title: Positioning of UWB and IMU with error state Kalman filter
Authors: Jayakusuma, Edo
Keywords: Engineering::Electrical and electronic engineering::Applications of electronics
Issue Date: 2021
Publisher: Nanyang Technological University
Source: Jayakusuma, E. (2021). Positioning of UWB and IMU with error state Kalman filter. Final Year Project (FYP), Nanyang Technological University, Singapore. https://hdl.handle.net/10356/150354
Abstract: Indoor-positioning has proved to be an important problem in recent years, because of the increasing urban environment complexity and unavailability of GPS indoors. Recently Ultra Wideband (UWB) have proved to be an emerging promising technology to solve this problem. As Ultra Wideband (UWB) have high latency, previous work have incorporated Inertial Measurement Unit (IMU) sensor fusion to mitigate this shortcoming. Most popular fusion algorithm to use is based on Extended Kalman Filter, and Unscented Kalman Filter. However, long duration of usage will have some reduced precision, as there are bias drifts for accelerometer and gyroscope measurement. A recent paper by Liu Et Al. is addressing this problem. Which uses Error State Kalman Filter to do online bias estimation of the IMU. Other than that, tightly coupled attitude determination is available with ESKF, with no need to rely on other algorithms like Mahony Filter for example. The author wants to reenact the experiment by Liu et al. to demonstrate the feasibility of such methods.
URI: https://hdl.handle.net/10356/150354
Fulltext Permission: restricted
Fulltext Availability: With Fulltext
Appears in Collections:EEE Student Reports (FYP/IA/PA/PI)

Files in This Item:
File Description SizeFormat 
Final_FYP_report.pdf
  Restricted Access
1.07 MBAdobe PDFView/Open

Page view(s)

137
Updated on May 18, 2022

Download(s)

8
Updated on May 18, 2022

Google ScholarTM

Check

Items in DR-NTU are protected by copyright, with all rights reserved, unless otherwise indicated.